Path to Action – Closing the Gap to Make Patient Blood Management the New Norm(al) as Viewed by Implementors in Diverse Countries


 Background: Patient Blood Management (PBM) is an evidence-based approach in surgery and emergency care which aims to minimize the risk for blood loss and the need for blood replacement for each patient through a coordinated multidisciplinary care process before, during, and after surgery. In combination with blood loss, anemia is the main driver for transfusion and an independent risk factor for adverse outcomes including morbidity and mortality. Hence, identifying and correcting anemia as well as minimizing blood loss are important pillars of PBM. Evidence demonstrates that PBM significantly improves outcomes and safety while reducing cost by macroeconomic magnitudes. Despite its huge potential to improve healthcare systems, PBM is not yet adopted broadly. The aim of this study is to analyze the collective experiences of a diverse group of PBM implementors across countries reflecting different healthcare contexts and to use these experiences to develop a guidance for initiating and orchestrating PBM implementation for stakeholders from diverse professional backgrounds.Methods: Semi-structured interviews were conducted with 1-4 PBM implementors from 12 countries in Asia, Latin America, Australia, Central and Eastern Europe, the Middle East, and Africa. Responses reflecting the drivers, barriers, measures, and stakeholders regarding the implementation of PBM were summarized per country, and key observations extracted. By clustering the levels of intervention for PBM implementation, a PBM implementation framework was created and populated.Results: A set of PBM implementation measures were extracted from the interviews with the implementors. Most of these measures relate to one of six levels of implementation including government, healthcare providers, funding, research, training/education, and patients/public. Essential cross-level measures are multi-stakeholder communication and collaboration.Conclusion: This implementation framework helps to decompose the complexity of PBM implementation into concrete measures on each implementation level. It provides guidance for diverse stakeholders to independently initiate and develop strategies to make PBM a national standard of care, thus closing current practice gaps and matching this unmet public health need.

However, despite compelling evidence and ongoing WHO policy drive, practical guidance for healthcare providers (HCPs) and national authorities (21,(25)(26)(27) and clinical guidelines and recommendations across numerous specialties and national health systems (22,(28)(29)(30)(31)(32)(33)(34)(35)(36), implementation of Patient Blood Management is far behind the expectations for good and safe clinical practices. Ignoring the cumulative evidence puts life, well-being and safety of millions of hospitalized patients at risk. Delaying Patient Blood Management implementation also means that healthcare systems forego savings of macro-economic magnitudes from a system-wide implementation of Patient Blood Management (20). This is even more alarming in countries striving towards Universal Healthcare Coverage and with severe resource constraints. In 2016, Eichbaum et al compared the Patient Blood Management implementation status in four countries using a six-questions survey and observed considerable variation between countries driven both by differences in health contexts and disparities in resources (37). They concluded that comparing Patient Blood Management strategies across low-, middle-, and high-income countries should foster mutual learning and implementing innovative, evidence-based strategies for improvement.
By gathering the experiences of a diverse group of implementors in Patient Blood Management across countries with different economic and healthcare contexts, and with track records of success in implementing Patient Blood Management regionally or locally, we aimed to identify the status-quo and approach of the implementation in each of the surveyed countries, and the drivers, barriers, measures, and stakeholders regarding the implementation of Patient Blood Management.
Synthesis and analysis of this information serves to provide an implementation framework for Patient Blood Management, including structural and procedural measures at various levels of intervention to improve patient outcomes at a large scale.

Methods
Semi-structured interviews mostly lasting 45-60 minutes were conducted between November 2019 and May 2020 with a multidisciplinary group of 36 Patient Blood Management implementors leading the implementation of Patient Blood Management in their respective environment. The twelve countries from Latin America, Central and Eastern Europe, Asia-Paci c, Middle East and Africa re ected experiences from high income and Low-and Middle-Income Countries (LMIC), different levels and types of healthcare resources and system (national/ private funders, public / private providers), and different developmental stages of Patient Blood Management (from early stage to fully integrated on policy level). The interviewees differed by clinical discipline (e.g., hematologists, anesthesiologists, surgeons) and perspectives (e.g., clinical specialists, blood bank, policy, Patient Blood Management coordinator, industry). The interviews followed a newly developed questionnaire (Additional File 1). One question required rating of prede ned barriers between 0 (not important) and 4 (very important), all other nine questions were formulated open without prompting speci c answers. The survey was rst piloted with eleven interviewees. Most interviews were conducted via web-communication (GoToMeeting™) by a single interviewer (AP Holtorf, Dr. rer. nat, female) in English language, two interviews were conducted by a second quali ed male interviewer in Chinese language after detailed brie ng by the main interviewer. The interview questionnaire was provided to the interviewees at least one week before the interviews. During the interviews, the interviewees consented to note-taking, recording, and publication of the results. The notes were revised using the recordings and the interviewees had the opportunity to review, correct or complement their initial responses.
The synthesis and analysis to provide the implementation framework followed six steps: (1) Responses per country were consolidated in a structured summary document (from two to four interviews per country except for Switzerland with one). (2) Responses from all countries regarding status-quo, approach of the implementation, and (3) drivers, barriers, measures, and stakeholders for Patient Blood Management were categorized and transferred in an electronic spreadsheet. (4) The categorized responses from step three were ranked for the frequency of mentions. (5) Accelerating and inhibiting factors were pooled and translated into implementation measures. (6) The measures were grouped by the interventional levels (policy/government, funding, research, healthcare provision, training/education, and public / patients). Steps 1 to 5 were conducted by the main interviewer.

Demographics
Thirty-six Patient Blood Management implementors, named "Patient Blood Management Implementation Group" with 15 women and 21 men from 12 countries, were interviewed following eleven pilot interviews (total of 47). The respective perspectives are depicted in Table 1.
Drivers for the implementation of Patient Blood Management -Why should it be done?
In Figure 1, the most prominent of the eleven drivers spontaneously mentioned during the interviews were patient outcomes (26 mentions), cost savings (23 mentions), preventing or better dealing with blood shortages (16 mentions from KOR, TUR, MEX, CHN, BRA), improving patient safety or reducing complications (15 mentions from BRA, CHN, LBN, KOR, SAU, TUR). Several experts mentioned national policy (8), education and awareness (concerning the risks of transfusion and bene ts of Patient Blood Management) (7), and a quality assurance system (6).
Shorter length of hospital stays, better use of resources, and reduction of waste were only mentioned once each. Patient demand was considered to become a driver once the risks related to transfusion and the bene ts Patient Blood Management were recognized more broadly in the general population.

Barriers for the Implementation of Patient Blood Management
Except for Australia, where Patient Blood Management is already widely adopted into practice, the need to change work practice was rated as the most prominent barrier for the implementation of Patient Blood Management as shown in Table 3. The need for collaboration and communication was rated equally important across the countries, followed by the lack of experience with Patient Blood Management, thefeasibility to integrate Patient Blood Management into the current processes, and strong belief in transfusion.

Accelerators and inhibitors for the implementation of Patient Blood Management
The responses for factors accelerating or supporting Patient Blood Management implementation fell into 24 categories as shown in Figure 2. Generation of local data and evidence, education and training for Patient Blood Management, a national Patient Blood Management policy, and strong thought leadership, were the most frequently mentioned factors. Blood scarcity, funding, awareness of transfusion risks, incentives for Patient Blood Management engagement, belief and commitment of care personnel, and quality assurance obligation were also frequently mentioned. During the nal six interviews between February and May 2020, the COVID-19 pandemic was newly mentioned as potential accelerator due to increased blood scarcity and potential blood safety issues.
The inhibitors or delaying factors fell into 22 categories (see Figure 2). with the most frequently mentioned being low awareness, no funding for set-up cost, education gaps, and stickiness of the old practice (even stronger if combined with the responses for the closely related resistance against change), lack of interdisciplinary commitment, and resistance against change.

Stakeholders
Sixty-three percent of the interviewees (29 of 46) included policy makers (National Health Council, Ministry of Health (MoH), etc.) as important stakeholders in Patient Blood Management implementation. As shown in Figure 3, the majority also listed either specialists in general (22), or speci c specialists (12 x anesthesiologists, 7 x hematologists, 5 x surgeons), 35% (16 of 46) included the hospital management. Other stakeholders (professional societies, national or regional blood banks, payers, nursing staff, enthusiastic champions, hospital pharmacists, patients/patient organizations, pharmaceutical companies, researchers/academics, hospital champion, general practitioners (GP) were mentioned less frequently or only in other parts of the interview (Medical schools, nongovernmental organizations, or the public at large).
The Patient Blood Management Implementation-Matrix After translating accelerators and inhibitors into actionable measures, six levels for intervention were identi ed: government/policy, funding, research, HCPs, education/training, and public/patients. Each of the six levels contributes with level-speci c relevant measures to the implementation of Patient Blood Management as re ected in Table 4  HCPs: Implementors expected HCPs to bene t from Patient Blood Management through improved outcomes at reduced cost. The identi cation of local champions and allies, the securing of funding, information technology (IT) infrastructure and support to enable Patient Blood Management data collection, reporting and benchmarking was deemed equally necessary as establishing multiprofessional teams, Patient Blood Management committees, program coordinators and nurses. Many preferred a piloting approach ("harvest low hanging fruit"), accompanied by the development of internal capability, to gain practical experience and to optimize the Patient Blood Management processes in the local context. The stepwise approach also included developing Patient Blood Management standard operating procedures, de ning key performance indicators, and measuring outcomes. Electronic clinical decision support systems for administering transfusions were deemed effective, also if combined with systems to incentivize, 'nudge', and reward the progression towards Patient Blood Management.
Training and Education: Implementors agreed that clinical Patient Blood Management knowledge and skills must be embedded in both under-and postgraduate education (curricula in medical schools, accredited continuous medical education (CME), Patient Blood Management academies, and e-learning-and information-platforms).
Research: Generating local evidence (prove of outcomes and cost-effectiveness in the local context at local cost structures) was deemed essential by the implementors, but also Patient Blood Management practice improvement through new research. Again, supporting IT architecture was seen essential for capturing and linking local outcomes data (morbidity, mortality, blood utilization/transfusion, cost). International exchange programs for Patient Blood Management research were reported to foster international collaboration and best practice development. Patients: Implementors mentioned multiple bene ts for patients including improved patient satisfaction. Successful implementation of Patient Blood Management would support shared clinical decision making, and individualized treatment plans, nally leading to increased demand for Patient Blood Management and decreased demand for blood. Some implementors saw patient advocates as potential ambassadors for Patient Blood Management, reaching out to government and funding level stakeholders as well as liaising with the patient communities concerned.

Discussion
The Challenge The implementation of Patient Blood Management is hampered by barriers mostly related to the di culty of changing traditional "physician's attitudes" towards transfusion (40)  indicates, that blood utilization is rather driven by culture and behavior than evidence (42,43,(53)(54)(55)).
An essential challenge in replacing this long-standing, well-organized, product-centered culture by a patient-centered treatment model is that most diverse stakeholders need to communicate, collaborate and overcome the complexity of the Patient Blood Management implementation process. This starts with their speci c contribution to the systemic implementation as exempli ed in Table 5 and presented in more detail in the additional online material (Additional File 2).

Using the Implementation Matrix to Develop Patient Blood Management Strategies
Unless translated into the daily routine and organizational culture, evidence is of limited value (56). To bridge the gap and effectuate the necessary culture change, it is essential to understand the drivers and barriers for Patient Blood Management as well as the stakeholders' roles and responsibilities. The Patient Blood Management-implementation matrix, as derived from the interviews, guides Patient Blood Management implementors in systematically identifying effective measures for Patient Blood Management implementation depending on the economic and healthcare context in their country. These measures will be discussed in more detail along six implementation levels.

Government Level
The focus of Patient Blood Management is to improve patient outcomes by managing and preserving the patient's hematopoietic system in surgical and medical settings. Corollaries of Patient Blood Management are decreased demand for blood products and signi cant reduction in average length of hospital stay. In Western Australia, hospital stays were reduced by almost 70,000 days over 5 years (20). Suchlike improvement enhances capacity of care and consequently, patient access, and resource utilization. Reduced morbidity, mortality, and improved patient safety related to Patient Blood Management are likely to increase life-expectancy, healthrelated quality of life, and national productivity. Massive savings due to Patient Blood Management allow for better allocation of scarce resources, thus increasing productivity of the healthcare sector. The multiple-win advantage of Patient Blood Management (57) support national healthcare priorities such as better equity, access, and affordability. This should motivate national policy makers to sharpen the national policies by prioritizing Patient Blood Management.
A national Patient Blood Management policy, as suggested by some of the implementors, needs to address the broad scope of Patient Blood Management with a bundle of measures as described earlier (58). Introducing Patient Blood Management through local Patient Blood Management pilot programs can happen more rapidly than a full national policy and program, and may serve as a prototype proving feasibility, success, and effectiveness in the local context.
Structural changes on government level usually require long time. One implementor stated "it takes more than seven years to introduce a policy in our country". Creating a sense of urgency through multiple stimuli can help to overcome the inertness for introducing a new medical model perceived as being complex (15).

Healthcare Provider Level
Patient Blood Management offers the rare opportunity to improve patient outcomes while reducing resource utilization and cost (20,59,60). The HCP related measures reported by the implementors start with the identi cation of local champions and allies from clinical and non-clinical departments but also encompass establishing multi-disciplinary teams including IT and administration, Patient Blood Management committees, program coordinators and other dedicated staff, and securing of funding, reinforcing the recommendations by previous experts (25,58).
Introducing Patient Blood Management practically via piloting accompanied by internal capability building through training and gaining practical experience, also is aligned with published approaches to change (61,62). Developing Patient Blood Management standard operating procedures helps tailoring the general guidance to the local context, and electronic transfusion decision support systems can effectively reduce transfusion rate and index in the daily routine(63,64) and serve as a 'nudging' mechanism. 'Nudging' denotes "non-regulatory and non-monetary interventions for changing behavior that steer people in a particular direction while preserving their freedom of choice" (65,66). This includes automated or targeted reminders, individual performance reviews based on local data collection and analysis, or Patient Blood Management dashboards as reported elsewhere (67

Funder Level
Public funders may bene t from Patient Blood Management through reduced average length of hospital stay and lower resource consumption, resulting in cost containment and better resource use. Private funders may expect higher pro tability, in particular with diagnosis related groups (DRG) or value-based reimbursement systems (e.g., accountable care): in DRGs with high anemia prevalence and potentially high blood loss such as obstetrics, cardiovascular surgery or oncology, the total cost per episode of care have shown to decrease over time, thus leading to reduced tariffs (69). For Germany, overall yearly cost-savings with elective surgery were calculated to be €1,029 million -almost 1.58% of the total national hospital budget (70).
Even in fee-for-service settings, funders may bene t from Patient Blood Management: currently, they might reimburse hospitals for the number of transfusions administered, while patients pay for their anemia treatment out-of-pocket. Where transparent, implementors in the interviews reported increasing cost of blood components (per unit) due to increasing measures for quality and safety testing. Once funders stop to incentivize transfusion and begin incentivizing (pre-operative) anemia management as an essential part of Patient Blood Management, they foster better outcomes, fewer complications, and shorter hospital stays, thus reducing the overall reimbursement cost per episode of care. This principle holds even true in healthcare systems where allogeneic blood products are covered by national funds and are considered 'free', because the cost of quality assurance and administering these blood products is a multifold of their actual acquisition cost and therefore represents a substantial cost volume for the hospital and consequently for the funder (71,72).
Given the documented savings potential with Patient Blood Management (20,70,(73)(74)(75)(76), it should be a priority for implementors to inform, educate and engage funders on this important issue. Following the example of the German health insurance BARMER (69), insurers might even help underpinning the Patient Blood Management value using their own data to demonstrate savings with improved outcomes.

Patient Level
According to the implementors, Patient Blood Management and its bene ts are largely unknown to patients, despite being the 'big winners' from Patient Blood Management with signi cantly improved clinical outcomes, safety, and reduced average length of hospital stay. Patients usually seek medical treatment based on a proper diagnosis and expect 'their problem to be xed' with safe and effective medical or surgical interventions. Unless being informed by their treating physician or alerted by credible public information, they would not know that Patient Blood Management improves their chances for earlier discharge from hospital and reduces their risk for hospital acquired infection or even mortality. Patient advocates could contribute by creating Patient Blood Management awareness, but also by educating, and defending patients' rights. Collaborating and likewise, supporting national campaigns to emphasize safety and the bene cial outcomes of Patient Blood Management, could foster shared clinical decision making and informed consent. Some implementors also saw the potential for patient advocates to approach funders to incentivize and support Patient Blood Management.
However, one implementor apprehended that entering the public domain too early might carry the risk of creating demand before physicians would be su ciently familiar with Patient Blood Management and its bene t. Another implementor cautioned, that too much information on transfusion risks may negatively impact on the willingness to donate blood. Improving patient outcomes and using donated blood more effectively should always remain the priority objective of Patient Blood Management. Involvement of patients or patient advocates should happen thoroughly and be planned within the country culture and context. However, the aim to involve patients more in their own care(77), the strive for 'Person-centered healthcare' (78), and the priority of increased patient safety (79)(80)(81) conforms to physicians' obligations towards educating and informing patients about all risks and bene ts of available treatment options. Medico-legal experts increasingly caution that widespread disregard of transfusion associated risks for adverse outcomes may result in litigation against those physicians and specialists (82). Informing the public and the patients in collaboration with patient advocacy groups can be a powerful element of the Patient Blood Management implementation strategy. Engaging the public and patients will not only result in more demand for Patient Blood Management as best practice but also improve patient satisfaction and foster participatory medicine.

Guided Implementation
In some of the countries described in this survey, Patient Blood Management was implemented simultaneously from bottom-up (e.g., from a department level or hospital/clinical level) and top-down (driven by policy and/or hospital administrative leadership) (see Table   2) with large variation in the closeness of the interaction between policy and operational levels. In other countries, implementation still progresses just through the bottom-up pathway, predominantly initiated, and led by individuals or small groups with different clinical background or innovation managers. To effectively coordinate and execute a statewide or even national implementation project across six diverse but interdependent layers requires governance, a pivotal element for bundling the power, control, bureaucracy, organization and legislative initiative (20,26,83 The experiences and expectations of the implementors con rmed how important the implementation into the local healthcare and cultural context and alignment with the local / national healthcare priorities and funding situation is. Implementation success depends on good change processes; pushback from the old transfusion paradigm due to ignorance and con icting incentives needs to be overcome, and the Patient Blood Management paradigm must be anchored in the healthcare delivery culture. Kotter's model for managing change embraces eight essential accelerators: establishing a sense of urgency, creating a guiding coalition, developing a change vision, communicating the vision for buy-in, empowering broad-based action, generating short-term wins, never letting up, and incorporating changes into the culture (84,85). Concurrent action, well adapted to the local context, across all eight change accelerators while rapidly building a network of change agents should maximize its adoption and impact (85).
Adaptation to the local context depends on the access to the key stakeholders and in uencers within the own healthcare environment.
Implementors need to identify the stakeholders in implementing Patient Blood Management and understand what motivates each of them to support, engage, or contribute (25)(26)(27).
The full implementation matrix (Additional File 2) may serve as a guidance in planning, even if starting with a small pilot. In creating the own path with clear aims, the user should, on each level, assess what works (best) in the local context, when, and which stakeholders should be involved (following Realist Evaluation approach (86)).

Limitations
The following limitations should be considered for this research. The selected countries cannot be fully representative for all countries and healthcare systems across the world. However, they were from ve continents and represented healthcare systems of high or lower income, and the interviewees were professionals leading and/or promoting Patient Blood Management in the healthcare sector of their respective environment. The impact of the various implementation measures across six levels could not be determined. Once Patient Blood Management will be established in more countries and healthcare systems, certain key performance indicators might be linked to speci c measures and rated, thus showing their relative importance.

Conclusion
With the objective of learning from the practical experiences with the implementation of Patient Blood Management, structured interviews were conducted with a multi-disciplinary group of Patient Blood Management implementors in twelve countries re ecting initial, advanced, and full level of implementation. Ethics approval and consent to participate No ethical approval was sought because the research did not entail systematic collection or analysis of data in which human beings are exposed to manipulation, intervention, or observation (WHO Manual (Section XV.2)). All interviewees were professionals who were informed about and agreed to the purpose of the interviews before and at the beginning of the interview. No personal data beyond the interviews with the implementors were used. The interview guide underwent internal compliance review by the pharmaceutical company interviewee in each of the participating countries.

Consent for publication
All interviewees consented to and supported the use of the information for this research and publication. All main authors have consented to the publication.

Funding
The research and analysis for this manuscript was funded by Vifor Pharma AG, Switzerland. None of the interviewees received any type of compensation for their time and input.
Authors' contributions APH and AH conceptualized the research plans. DRS plaid a critical role in testing and improving the interview ow. APH conducted and evaluated all interviews. APH, AH and DRS were major contributors in writing the manuscript. All authors read and approved the nal manuscript.